Hi, Lloyd,
> On Nov 19, 2024, at 3:58 PM, lloyd.w...@yahoo.co.uk
> <lloyd.wood=40yahoo.co...@dmarc.ietf.org> wrote:
>
> ...
> When UDP-Lite (RFC3828) was proposed, it was eventually shunted to its own IP
> protocol number, rather than risk messing up existing UDP implementations and
> semantics for even a very minor change. I gather that move to a different
> port was done very late in the process.
It was proposed in the very first WG draft:
https://datatracker.ietf.org/doc/html/draft-ietf-tsvwg-udp-lite-00
Therefore, this draft proposes to allocate a new protocol identifier
for UDP Lite.
Even the Lars-Åke Lazon’s MS thesis from 1999 noted that it was not backwards
compatible with legacy endpoint (in terms used in the UDP options draft).
That’’s quite different from one of the original design goals of UDP options
(see section 6 f the current doc).
Then again, that’s also probably why Lite took only 2 years from draft to RFC,
and options are approaching 10.
Joe
_______________________________________________
Int-area mailing list -- int-area@ietf.org
To unsubscribe send an email to int-area-le...@ietf.org
